🧪 TrigVideoCapture Toolbox

A Python-based toolbox for high-speed video acquisition with FLIR / Blackfly USB3 cameras using the Spinnaker SDK / PySpin.

Initially developed for neuroscience experiments, this toolbox supports live camera display, video acquisition, frame timestamp logging, and trial/stimulus metadata logging from either NI DAQmx or an Arduino serial decoder.

---

✅ Features

• Real-time video display
• Video recording from FLIR / Blackfly cameras
• Support for 1 or 2 cameras with the flexible Arduino script
• Timestamp logging for each acquired frame
• Metadata logging to CSV
• Arduino-based serial decoding of trial/stimulus signals
• Optional NI DAQmx-based analog acquisition
• Auto-generated filenames
• Optional split-recording mode (WIP)

---

🚀 Usage Modes

Script	Input Device	Camera Support	Notes
Display_Record_Arduino_1or2Cam_AIO.py	Arduino serial decoder	1 or 2 cameras	Recommended flexible Arduino version
Display_Record_Arduino_AIO.py	Arduino serial decoder	2 cameras	Older fixed dual-camera version
Display_Record_DAQmx_AIO.py	NI DAQmx	Depends on script version	Reads analog input continuously

---

🔌 Feldman Lab — Minimal Hardware Connections

Camera acquisition

Source / Device	Destination	Notes
Camera USB 3.0	Acquisition computer	Plug directly into a USB 3.0 port for full bandwidth
Second camera USB 3.0	Acquisition computer	Only needed for two-camera setup

For a single-camera setup, connect only one camera.

For a two-camera setup, connect both cameras before starting the Python script.

---

Arduino trial/stimulus metadata

Source / Device	Destination	Notes
Vout from Adafruit MCP4725 DAC	Arduino A0	Analog packet encoding trial/stimulus values
TDT Start Signal	Arduino digital pin 2	Rising edge starts packet decoding
TDT / DAC GND	Arduino GND	Required shared ground
Arduino USB	Acquisition computer	Used for serial communication with Python

The Arduino decodes the analog packet and sends lines such as:

TRIAL:12,STIM:3

The Python script reads these values through the serial port and stores them in the frame metadata CSV.

---

Optional LED illumination control

Source / Device	Destination	Notes
Arduino D9	LED 1 driver / MOSFET input	PWM output for LED 1
Arduino D10	LED 2 driver / MOSFET input	PWM output for LED 2
Arduino A2	Potentiometer 1 wiper	Controls LED 1 intensity
Arduino A4	Potentiometer 2 wiper	Controls LED 2 intensity
Arduino 5V / GND	Potentiometer side pins	Potentiometer reference voltage
Arduino GND	LED driver ground	Required shared ground

The Arduino script also uses a door interlock:

Source / Device	Destination	Notes
Door switch	Arduino D8 to GND	Uses INPUT_PULLUP

If no door switch is used during testing, connect:

Arduino D8 → GND

Otherwise the Arduino will interpret the door as open and keep the LEDs off.

---

🧰 Requirements

• Python 3.10
• Conda recommended
• FLIR Spinnaker SDK
• PySpin, installed from the Spinnaker SDK Python wheel
• OpenCV
• NumPy
• tkinter, usually bundled with Python
• For Arduino version: pyserial
• For DAQ version: nidaqmx

---

📦 Installation

This has only been tested on Windows machines, but should in principle run on other OS

You first need to install Spinnaker (https://www.teledynevisionsolutions.com/support/support-center/software-firmware-downloads/iis/spinnaker-sdk-download/spinnaker-sdk--download-files/) Currently( June 2026) the download includes an .exe to install Spinnaker and some compressed wheel files that allows to install SDK for python (e.g. spinnaker_python-3.2.0.65-cp310-cp310-win_amd64.whl)

During the installation of Spinnaker, installing only the spinView program is sufficicent, it will allow you to choose the camera parameters (exposure, frame rate etc...)

Create and activate a conda environment. ( I use miniconda)

On Windows, open an Anaconda Prompt as administrator: and type all of the following:

conda create -n TrigVideoCapture python=3.10 -y
conda activate TrigVideoCapture

python -m pip install --upgrade pip setuptools wheel
python -m pip install numpy==1.23.5 pandas pyserial opencv-python
Install the Spinnaker Python wheel.

install PySpin, the FLIR SDK, from the wheeler . it needs to match your OS (win_amd64) , python in the conda (3.10), and normally version of Spinnaker just installed (3.2.0.65)0

Example: Change the path to wherever your wheel is.

pip install "C:\...\USER\download\spinnaker_python-3.2.0.65-cp310-cp310-win_amd64.whl"

Test PySpin installation:

python -c "import PySpin; print('PySpin file:', PySpin.__file__); print('Has System:', hasattr(PySpin, 'System')); print(PySpin.System)"

Then install the remaining Python requirements:

pip install -r requirements.txt

---

⚙️ Camera Configuration

Camera parameters such as:

• frame rate
• exposure time
• gain
• image size / ROI
• trigger mode, if used

should be configured in the Spinnaker SDK GUI before running the acquisition script.

The Python script records video at the camera framerate configured in Spinnaker.

---

🚀 Running the Toolbox

Activate the environment and go in the folder where this github code lives:

conda activate TrigVideoCapture
cd PathToTheToolbox

Recommended Arduino version: 1 or 2 cameras

python Display_Record_Arduino_1or2Cam_AIO.py

This script automatically detects the number of connected cameras.

• If one camera is connected, it records one video.
• If two or more cameras are connected, it records the first two cameras.
• Each camera is saved as a separate video file.

Example output files:

MySession_cam0.avi
MySession_cam1.avi
MySession.csv
MySession_Analog.csv

For a single-camera setup, only MySession_cam0.avi is created.

---

Older Arduino dual-camera version

python Display_Record_Arduino_AIO.py

This version expects two cameras.

---

NI DAQmx version

python Display_Record_DAQmx_AIO.py

Use this version if metadata are acquired through an NI DAQ device instead of the Arduino serial decoder.

---

💾 Output Files

When the program starts, it will ask you to:

1. choose an output folder
2. enter a base filename

The toolbox then creates:

File	Description
BASE_NAME_cam0.avi	Video from camera 0
BASE_NAME_cam1.avi	Video from camera 1, if present
BASE_NAME.csv	Frame-by-frame metadata
BASE_NAME_Analog.csv	Arduino-decoded trial/stimulus log

The metadata CSV contains frame timestamps and trial/stimulus values decoded from Arduino serial communication.

---

🛑 Stopping Acquisition

To stop recording normally:

Press q while the video display window is active

For emergency stop:

Press Ctrl+C in the command prompt

---

⚠️ Notes

• Use a direct USB 3.0 connection for each camera when possible.
• Avoid USB hubs unless they are known to support the required bandwidth.
• Make sure the Arduino COM port and baudrate in the Python script match the Arduino.
• The Arduino script uses:

Serial.begin(115200);

so the Python baudrate should usually be:

SERIAL_BAUDRATE = 115200

• On Windows, check the Arduino COM port in Device Manager.
• The flexible 1-or-2-camera script includes safe overwrite checks for the actual output video names.